Abstract 469: Systematic analysis of circulating tumor DNA in melanoma patients to uncover mechanisms of resistance and disease clonality

• Published in: Cancer research (Chicago, Ill.) Ill.), 2016-07, Vol.76 (14_Supplement), p.469-469
• Main Authors: Gremel, Gabriela, Lee, Rebecca J., Girotti, Maria R., Garner, Grace, Mandal, Amit K., Wood, Sally, Fusi, Alberto, Valpione, Sara, Serra-Bellver, Patricio, Dhomen, Nathalie, Lorigan, Paul, Marais, Richard
• Format: Article
• Language: English
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2016-469

Recent advances in targeted and immunotherapies have unlocked potent treatment options for malignant melanoma patients. However, targeted therapies are associated with limited response durations and only a fraction of patients benefit from immunotherapies. The analysis of circulating tumour DNA (ctDNA) provides a powerful tool for the continuous assessment of treatment responses in melanoma patients. We have established a robust pipeline for the serial, prospective analysis of ctDNA in melanoma patients. By combining next generation sequencing (NGS)- and droplet digital PCR (ddPCR)-based approaches we make ideal use of currently available technologies. To date, plasma samples from over 120 individuals have been collected, with follow-up times surpassing one year in many cases. For routine monitoring of patients commencing BRAF-mutant targeted therapies, pre-treatment DNA, derived from tumour biopsy or plasma, is tested to confirm the exact BRAF mutation and rule out pre-existing resistance-associated mutations using a custom NGS sequencing panel. Timely follow-up throughout treatment is achieved by measuring mutant BRAF fractions in ctDNA using ddPCR. When increasing BRAF variant allele frequencies are detected, additional ddPCR-based profiling for NRAS mutations and NGS-based re-sequencing for other resistance-associated mutations is initiated. We demonstrate superiority of ctDNA monitoring over LDH measurements and present a strategy to detect emerging resistant disease ahead of clinical scans in a cost-effective, timesaving and low technology way. In addition to the basic characterization of BRAF-mutant cases undergoing targeted therapy, we have used the analysis of ctDNA to uncover clonally distinct tumour sub-populations resulting in complex responses to treatment. For instance, we monitored a case of metastatic vaginal mucosal melanoma undergoing sequential targeted, immuno- and chemotherapy. Despite the presence of a KIT mutation in the primary tumour, response to KIT inhibitor imatinib was mixed. In the absence of tumour biopsies, we used whole exome-wide NGS of ctDNA to study the underlying mechanisms. This strategy revealed a KIT-mutant tumour sub-clone that responded to imatinib and a second sub-clone including an SF3B1 mutation that did not respond. The sub-clones also responded differentially to immunotherapies, but both responded to chemotherapy. The analysis of ctDNA is a powerful approach to monitor responses to systemic therapies in melano